Robot Trajectory Retargeting With Precomputed Motion Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic devices face challenges in navigating complex environments with difficult terrain, unforeseen obstacles, and changing external conditions, requiring improved agility and dynamic control while efficiently managing computational resources.
Innovation Solution
The system employs a library of pre-computed 'template behaviors' such as running, walking, or jumping, which are adapted in real-time using information about the robot's kinematic state and environment, allowing for optimized trajectory planning and execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If detailed nonlinear trajectory optimizations are performed online in real-time, then the robot can adapt to complex environments and unexpected obstacles, but the computational resources and processing time required increase significantly
Solution Approach 1:
The patent pre-computes and stores a library of template trajectories offline that represent common movement patterns and behaviors. During online operation, the robot selects and adapts these pre-computed trajectories rather than performing full nonlinear optimizations in real-time, significantly reducing computational burden while maintaining adaptability to complex environments.
Solution Approach 2:
The trajectory planning is divided into two segments: offline pre-computation of template trajectories covering a range of scenarios, and online selection/adaptation of appropriate templates based on current environmental conditions. This segmentation allows detailed computations to be performed beforehand when computational resources are abundant, while online operations use only lightweight selection and adjustment.
2Speed
If the robot performs frequent online trajectory re-computations to respond to environmental changes, then the movement accuracy and responsiveness improve, but the computational time and energy consumption increase
Solution Approach 1:
Multiple trajectory templates are pre-computed offline to cover various environmental conditions and task requirements. When the robot encounters environmental changes, it selects from these pre-computed templates rather than computing new trajectories from scratch, enabling rapid response without extensive online computation.
Solution Approach 2:
The robot periodically selects and switches between pre-computed trajectory templates based on environmental feedback and task progress. This periodic selection approach maintains responsiveness to environmental changes while avoiding continuous heavy computation, as the heavy lifting was done offline during template generation.
3Power
If the robot uses simplified trajectory corrections online, then the computational intensity is reduced, but the precision and accuracy of movement may deteriorate
Solution Approach 1:
The template trajectories are pre-computed offline with high precision using detailed nonlinear optimizations. These pre-computed templates serve as accurate base trajectories, so online operations only require simple selection and minor adaptations rather than full re-optimization, maintaining movement precision while reducing online computational intensity.
Solution Approach 2:
The robot copies and adapts pre-computed trajectory templates rather than generating new trajectories online. The templates contain detailed, precise motion plans computed offline, and the online process involves selecting appropriate templates and making minor adjustments to fit current conditions, preserving precision without requiring intensive online computation.
Data Source
AI summary
Systems and methods for determining movement of a robot about an environment are provided. A computing system of the robot (i) receives information including a navigation target for the robot and a kinematic state of the robot; (ii) determines, based on the information and a trajectory target for the robot, a retargeted trajectory for the robot; (iii) determines, based on the retargeted trajectory, a centroidal trajectory for the robot and a kinematic trajectory for the robot consistent with the centroidal trajectory; and (iv) determines, based on the centroidal trajectory and the kinematic trajectory, a set of vectors having a vector for each of one or more joints of the robot.